Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 11-GEMMASCOPE-RES-16K
    6. 6109
    Prev
    Next
    INDEX
    Explanations

    references to clinical outcomes and statistical measurements in medical studies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    NameInMap
    -0.55
    ...</
    -0.51
     DEF
    -0.51
    aarrggbb
    -0.51
    יח
    -0.51
     hangers
    -0.51
     anything
    -0.51
     //$
    -0.50
    cdef
    -0.49
    elski
    -0.49
    POSITIVE LOGITS
    MigrationBuilder
    0.75
     تضيفلها
    0.75
     snippetHide
    0.74
    zieży
    0.60
     avoient
    0.57
     pères
    0.57
    reactstrap
    0.57
     étoit
    0.56
     μὲν
    0.56
     hasard
    0.55
    Activations Density 0.713%

    No Known Activations